Dear Siesta Users and Developers A fairly simple and perhaps naive question: How to find out the valence of a doping atom in calculated system?
I tried to utilize Mulliken, Hirshfeld, and Voronoi atomic populations analysis to find out the valence of doping atom, however these three schemes often give very different results. Moreover, the obtained values of the “dQatom” are far away from the valences that are widely accepted for given species. For example, typical valences of the Cobalt atom are: Co(2+) or Co(3+), i.e., 2 or 3 electrons are expected to be removed from the valence-shell s orbital or from the s and d orbitals. Nevertheless, using these three schemes, I typically get dQatom = +0.3….+0.6, i.e. the Co atom has lost 0.3-0.6 electrons, which seems extremely small and disagree with available literature data for some similar systems. These values are obtained taking into account the electrons on polarized 4Pp orbitals (by default), while I doubt whether these orbitals should be used in my analysis? I will be grateful for any clarification on how to correctly determine the valence of an atom. Thank you in advance for any advice. With kind regards, Dmitriy Muzychenko
-- SIESTA is supported by the Spanish Research Agency (AEI) and by the European H2020 MaX Centre of Excellence (http://www.max-centre.eu/)
